Peter E. Koveos

STOCHASTIC TIME-SERIES FORECASTS OF GOLD AND SILVER PRICES

ABSTRACT The task of forecasting commodity prices poses difficulties for many market participants. Recent gyrations in the markets have increased the rewards derived for accurate price forecasts. The purpose of this study is to arrive at forecasts for gold and silver spot prices and for the gold/silver ratio using the Box-Jenkins time series methodology. A Box-Jenkins time series model is an empirical representation of a stochastic process of an integrated autoregressive moving average form (ARIMA). Forecasts are obtained by computing the expected value of future observations that are implied by the model, conditional only on the past history of the series.

HEDGING EFFECTIVENESS IN PRECIOUS METALS

ABSTRACT The purpose of this paper is to develop the portfolio theory approach to hedging and to show its application to those holding a position in precious metals physicals. Simple and complex hedges are developed. A simple hedge relates a physical position in a single commodity to a single futures contract. A complex hedge permits purchasing of multiple futures when a position is held in one or more physical commodities. The portfolio hedging concept is developed for nearest 1-month, nearest 6-month, and nearest 12-month future contracts.